V THE ORGANIZATION OF BUSINESSES

This section sets out the basic legal rules on agency and liability and then ventures to suggest how, optimally, the Romans might have chosen to organize their business activities to take advantage of the background of these legal rules.

1. Representation and agency

New Roman">Although the Roman jurists developed a number of informal contracts such as sale, any contract which was in any way unusual or untypical would have to be entered into by stipulatio. That in turn required the presence of the contracting parties face to face. Communications being slow, there was a real restriction on how many contracts any individual could enter into in how many geographically dispersed locations; this strictly limited number was the necessary consequence of the fact that there was a strictly limited number of places in which a person could be at the same time. Accordingly, it became necessary to develop rules for representation, to allow the acts of one person to bind another. This was a breach with the strict principle of Roman law that obligations were personally binding only, and a contract gave rights only to the parties to it and to nobody else.

That rule did not apply, however, where one of the contracting parties was a slave or a child in the paternal power of his father.

All this changed with the development of legal remedies based on the peculium. The peculium was a fund of property granted by a paterfamilias to a person in his power, whether slave or free. It might be made up of any property: land, movables, businesses, other slaves. It remained the property of the paterfamilias and could be revoked by him, but in prac­tice the person in power had charge of it and could deal with it as if it were his own. There will be more to say about it in connexion with lim­itation of liability; for the present it is enough to note that the peculium allowed slaves and children to make themselves useful, since creditors were now accorded a right of action against the paterfamilias, albeit one restricted to the value of the peculium.

One consequence of the fact that a paterfamilias had power over his slaves and children was that any property and any rights acquired by them immediately vested in him. They served merely as channels by which contractual entitlements flowed to and vested in the paterfamil­ias. This was a feature of the most crucial importance where the pater­familias wished to enter into contracts at a distance. By sending a slave or a child there, he could himself acquire contractual rights in dealings entered into many miles from Rome. The importance of this can hardly be exaggerated.

In passing, it is important to clarify the scope of one piece of legisla­tion which might otherwise be thought to have restricted the role which sons could play in commerce. The senatus consultum Macedonianum, passed under the emperor Vespasian, prohibited loans to sons in paternal power, regardless of their age. According to Ulpian, the reason was that one borrower, Macedo, had found it necessary, in order to repay his creditors, to come into his money sooner rather than later and had there­fore murdered his paterfamilias (D. 14.6.i pr.). To discourage others from following this course, the senatus consultum prohibited lenders - even after the death of the paterfamilias — from recovering money lent to sons. A permanent defence against any claim for repayment was given, and this was expressly conceived as being for the purpose of penalizing the moneylender (Marcian, D. 12.6.40 pr.; Ulpian, D. 14.6.9.4). The market in such loans must therefore immediately have disappeared. Interpreted literally, this legislation would have brought all dealings with sons to a halt. But it was accepted that it did not apply where the paterfamilias had appointed his son to carry on trade or authorized him to do so from his peculium (both of these possibilities are discussed in the next section). It also applied only to loans and not to other transactions unless they were loans in disguise; and only if the creditor knew or ought to have known that he was lending to a son in power (Ulpian, D. 14.6.3 pr. and 3; Pomponius, D. 14.6.19).

2. Liability and limitation

The peculium

The praetor's innovation consisted in allowing creditors of slaves or dependent children to bring an actio de peculio against their paterfamilias.

This action was available only for acts done by a person under the power of a paterfamilias and was based mainly, but not purely, on rela­tions of status. It did not depend on any authority or task being dele­gated by the paterfamilias to the child or slave; nor did it depend on the paterfamilias knowing that the business was being carried on. Instead, it depended on the revocable grant of a fund of property to a person who stood in a certain relationship of status to the paterfamilias: commerce and status were therefore intertwined.

There are two important qualifications.

(1)   Actions based on the peculium were relevant only to the dealings of the child or slave with property or under contracts. Where a dependent member of the paterfamilias's family committed a civil wrong (a delict), such as theft, assault, or damage to property, the paterfamilias was auto­matically liable to pay the damages, although he could limit his liability by surrendering the wrongdoing child or slave to the person who had been wronged (‘noxal surrender'). But the peculium had nothing to do with cases of this sort.

(2)   The actio de peculio was appropriate only where the paterfamilias did not know of the business being carried on.

Actio institoria and actio exercitoria

The praetor introduced two further actions which were important in business dealings: they allowed creditors to sue the owner or principal of a business rather than the ‘agent' with whom they had actually dealt. (The terms ‘principal' and ‘agent' are used loosely.) The first, the ‘action on shipping' (actio exercitoria), was specific to shipping and allowed a claim against a ship owner (exercitor) for the acts of the ship's captain. The second, the ‘action on agency' (actio institoria), was generally available and allowed a claim against a person who had placed an ‘agent' or institor in charge of a business, for acts done by the agent in the course of the busi­ness. Two general points are worth noting. First, since these actions allowed customers to sue the owner of the business, whom they might never have seen before, they involved a breach of the fundamental prin­ciple that an obligation was strictly personal and bound only the person who had undertaken it. Second, the actual agent, the institor or captain, might be of any status: free, or in the power of a paterfamilias, or a slave. In each case the owner of the business had unlimited liability for the acts done by the agent in the course of the business. Where the agent was a slave - who could not be personally liable in contract - these actions were the customer's only remedy. On the other hand, if the agent was a free person, he or she would be personally bound by the contract and could be sued upon it, so these actions simply offered an additional remedy, an alternative defendant from whom the creditor could attempt to recover.

The circumstances in which liability under these two actions could arise were restricted by the terms of appointment (or praepositio) of the institor or captain: to use modern (though Latin-derived) terms, by the authority given by the principal to the agent. It was therefore open to the principal, by suitably restrictive drafting of the terms of appointment, to limit the circumstances in which liability came home to him at all.

Ulpian, in his discussion of the actio institoria, states that ‘not every transaction with an institor binds the person who appointed him, but it does so only if the contract was made on account of the business of which he was put in charge' (D. 14.3.5.11). There are other statements to the same effect. The principal could also reduce the scope of, or avoid, a liability that would otherwise arise, by giving express notice to parties contracting with his agent; there is a certain amount of discussion about what language such notices must be in, and how large and visible they must be. But it is a matter of contracting out of a liability which would otherwise exist. Plainly, the effectiveness of this limit on the principal's liability turns on how strictly the terms of the appointment are inter­preted. The Roman approach was a rather narrow one (D. 14.1.7; D. 14.3.13 pr.). Accordingly, a certain degree of protection was afforded by suitably tight drafting of the terms of the agent's or captain's appoint­ment. But if, for example, the agent or captain ran up huge losses within the terms of his appointment, the principal was without any protection. Where there was a significant degree of risk, it still made sense to rely on slaves or dependent children rather than independent labour: only that could offer a set financial limit on liability.

Ignorance - a good thing

If a slave or dependent child was running a business, whether the owner or paterfamilias was liable without limit under one of these actions or only up to the amount of the peculium depended on the paterfamilias's relationship with the business. The Digest texts make it plain that the paterfamilias was liable up to the amount in the peculium even if he had no idea what his slave or child was doing. On the other hand, a princi­pal was liable to the actio exercitoria or institoria only if he had actually placed the agent in charge of the business, or the captain in charge of the ship, and thereby shown his intention that it should be operated. In short, the paterfamilias could enjoy a financial limit on liability only where he did not appoint the slave or child to do anything, but remained at arm's length from the business. So the apparent advantage of limited liability offered by slaves and dependent children is only that: apparent. In fact, in order to gain that advantage the paterfamilias had to stay at arm’s length from the business and not appoint anyone to do anything. The law therefore encouraged a laissez-faire attitude on the part of the paterfamilias.

It may be that the explanation of this apparent conundrum lies in the servus vicarius. The peculium of a slave (a servus ordinarius) often included other slaves (known as servi vicarii); their peculia might equally contain slaves; and so ad infinitum. It is here that limitation of liability is able to play its true and effective role. The facts of one text from the Digest will serve as an example. Ulpian deals with the case where the shipowner (to use the term loosely) is not an independent person but is in the power of another person: he is a slave or dependent child. He therefore does not own the ship but simply has it in his peculium. He appoints a captain to run the ship. This involves one more layer than the basic case: not just shipowner and captain, but shipowner’s paterfamilias, shipowner and captain. The main consequence is that anybody wishing to bring the actio exercitoria based on what the captain does will have to bring it not against the shipowner but against his paterfamilias. Ulpian draws a dis­tinction: if the shipowner is carrying on that business by the will (volun­tas) of his paterfamilias, then the action can be brought against the paterfamilias without limitation. On the other hand, if the business is not being carried on by the will of the paterfamilias but by the will of the dependent shipowner, the paterfamilias is liable only up to the value in the shipowner’s peculium (D. 14.1.1.19).

Here we see the true significance of the rules described earlier. There is unlimited liability for acts done by the captain or manager within his terms of appointment, but this is true only at the level of the person who appointed him. The liability to which the ultimate owner of the ship or business is exposed is limited by the value of the peculium of the slave or dependant responsible for appointing the captain or business manager. That might be a dependant several degrees removed from the ultimate owner himself.

The legal sources therefore suggest that, depending on the level of risk involved in any particular enterprise, it would — ideally — be appropriate to have the enterprise managed not just by slaves or other dependent labour but by slaves who stood at some remove from the paterfamilias by reason of belonging to the peculium of one of his other slaves, and so offered the advantage that the paterfamilias had not himself appointed them to do anything, and might indeed have no knowledge of precisely what they were doing. When we consider how business was organized, at least by the wealthy, we should therefore be aware of the advantages which servi vicarii had to offer. The law presents significant advantages to businesses organized in the shape of a pyramid, with the paterfamilias at the apex and below him layers of managers and workers. This hier­archy also makes realistic the possibility that the owner of the slaves might have little idea what each was doing; and that itself justifies the central role played in the jurists' discussions by considerations about intention and will in determining the extent of the paterfamilias's liability.

Of course, the fact that this approach maximizes the advantages of one trader means that it also minimizes those of his trading partners. It follows that whether in fact a trader will be able to deal with his trading partners on such advantageous terms — or whether, for example, they will insist on his undertaking personal or unlimited liability in a transac­tion — will turn on their respective bargaining strengths. Accordingly, in reality the points mentioned so far would not be the only factors at work.

p. Independent labour

Slave labour was not the only labour available at Rome (Garnsey 1980). Independent employees acted as individuals and acquired rights solely for themselves. If they entered into a contract with somebody, there was no question of those contractual rights automatically vesting in their employer. This is quite different from the conception of agency with which we are now familiar, for example, in Scots and English law, where the acts of the agent can actually create contractual relations between the employer and the third party with whom the agent deals. In these systems the employer or ‘principal' does not have, for example, to have the agent assign his contractual rights to him. But in Roman law the employer did not enjoy this advantage. It is true that there would be no great difficulty in setting up another contract by which the independent employee agreed to transfer to his employer any rights he might acquire. Yet this would introduce an additional step into the situation which might cause problems: the only person who would be entitled to sue the third party would be the employee, who contracted with him; and if the employee became insolvent, it would be of little use to the employer to have a right of action against him. All these anxieties could be removed, however, by the simple expedient of making use of dependent labour: children and slaves. Probably until the early third century, there were advantages in using dependent labour for purposes of agency. Put simply, since the dependent agent was as a matter of law identified with the paterfamilias himself, to use such labour was in effect to cut out the middle-man.

None the less, the legal sources alone provide plenty of evidence of the employment of independent people, especially freedmen, sometimes doing the same jobs as they had done when they were previously slaves. Equally, there are faint signs of developments in the law making it easier to employ free people. From the second century ad, there was a slow but sure recognition that it was possible for one person to acquire rights through a free person, a procurator. This is said to have been more or less accepted in the case of acquiring possession at about the end of the first century ad (Neratius, D. 41.3.41). Certainly by the third century any doubts seem to have been overcome.

More generally, in the writings of the jurist Papinian at the end of the second century or beginning of the third, there are clear indications that actions might be available, on account of dealings carried out by a proc­urator, both for and against his employer. This does not mean that inde­pendent agents had not been used before. But it is probably right to detect in this development that a need had been perceived and was being gradually addressed by the law, to make remedies available between the parties genuinely interested in a transaction, and not to make the efficacy of remedies dependent on contingencies such as the solvency of the procurator. It may also be right to see in this development signs of what is sometimes called the ‘juridification’ of relationships: the incor­poration within a legal framework of relations which had previously been based purely on amicitia or officium. However that may be, what is clear is that in a system which had wholeheartedly favoured the employ­ment of dependent labour, the balance was being redressed to make the employment of independent labour more straightforward.

4. Partnership

One of the consensual contracts developed by Roman law was that of partnership (societas). It is quite clear that the background to this was far from commercial: the earliest precursor of partnership was an arrange­ment under which co-heirs of an inheritance continued to own and

107 administer it in common. This seems to have left its mark on the law of societas, as it eventually developed, since one form of partnership was a partnership ‘of all property' (omnium bonorum) which seems most unsuit­able for commercial purposes.

None the less, it was possible to enter into a partnership for one line of business only or for a single transaction, and this would bring the advantages of combining the skills and expertise, and not least the capital, of the partners. Apart from ordinary commercial ventures, a typical instance of partnership in the late republic was that of tax col­lectors (publicani); to some extent this was governed by special rules (Buckland 1963: 513).

The partners were free to make whatever terms they would for sharing profits and losses, except that it was not permitted to have a partner who shared only in the losses and not in the profits. In the absence of other provision, shares in profits and losses were equal. A Dacian document of ad 167 sets out terms for a partnership in — so far as its fragmentary state reveals — some detail. Oddly enough, it then goes on to confirm them by means of a stipulatio. Since the document con­tains only a stipulatio by one of the partners undertaking obligations to the other, it seems most likely that another, reciprocal version of the doc­ument would also have been produced. The alternative is that the doc­ument simply betrays a misunderstanding of the law (FIRA 3.157).

To modern eyes the most striking feature of the Roman partnership is that the only people on whom it normally had any effect were the part­ners. To the outside world its existence was of no significance: so if one partner entered into a contract to sell goods, he alone was bound by the contract. The partnership agreement mattered only to him and his partner, since under it he might be obliged to communicate to his partner some of the benefits he derived from the contract. The only exception to this rule would be if the other party to the contract (in this example the buyer) could show that the partner with whom he had dealt had been acting as the agent (institor) of the other: in that case, as explained already, he could sue the ‘principal' partner, even though he had had no dealings directly with him. For these reasons, although the possibility of entering into a partnership brought advantages in terms of sharing costs and resources, it brought no particular benefit as a means of structuring a business.

5. Some conclusions on business organization

The development of the law of agency through procurators certainly made it easier to make use of independent labour within the organiza­tion of a business. Equally, the adoption of a strict approach to inter­preting the terms on which an institor or captain had been appointed made it possible with appropriate drafting to make use of them for specific highly specialized purposes, without giving too much of a hostage to fortune. But there was still no substitute for making use of a slave or dependant at critical junctures of the business, particularly in high-risk activities, because only they could offer the advantage of a lim­itation to the value of their peculium.

The legal advantages of a laissez-faire approach to business fit rather well with what is known about social attitudes to trade. There is plenty of evidence in the Digest and elsewhere of a healthy disdain for involve­ment in trade. There were also restrictions on the extent to which sena­tors could engage in trade (Talbert 1984: 45-6). These are of course issues quite different from benefiting, or drawing the profits, from trade: the aristocracy may not itself have traded, but it certainly took the profits from trade (Plutarch, Cato maior21.6—7; Whittaker 1993: 58). That picture — of keeping a safe distance from trade — is complemented by the con­clusions drawn here from the legal sources: not simply (as is well known) that it was possible to trade by means of intermediaries, but also that there were positive advantages to be had by reducing one's level of involvement and knowledge of the business and to confining oneself to enjoying profits vicariously. We can draw some comfort from the fact that social attitudes and legal rules point towards the same kind of organization of business, along laissez-faire lines. This suggests that the law is not operating here in its own remote world of isolated rules but is firmly anchored in the realities of Roman life and the demands at least of its well-heeled citizens.